Trace Driven Analysis of the Long Term Evolution of Gnutella Peer-to-Peer Traffic

Size: px
Start display at page:

Download "Trace Driven Analysis of the Long Term Evolution of Gnutella Peer-to-Peer Traffic"

Transcription

1 Trace Driven Analysis of the Long Term Evolution of Gnutella Peer-to-Peer Traffic William Acosta and Surendar Chandra University of Notre Dame, Notre Dame IN, 46556, USA Abstract. Peer-to-Peer (P2P) applications, such as Gnutella, are evolving to address some of the observed performance issues. In this paper, we analyze Gnutella behavior in 23, 25, and 26. During this time, the protocol evolved from v.4 to v.6 to address problems with overhead of overlay maintenance and query traffic bandwidth. The goal of this paper is to understand whether the newer protocols address the prior concerns. We observe that the new architecture alleviated the bandwidth consumption for low capacity peers while increasing the bandwidth consumption at high capacity peers. We measured a decrease in incoming query rate. However, highly connected ultra-peers must maintain many connections to which they forward all queries thereby increasing the outgoing query traffic. We also show that these changes have not significantly improved search performance. The effective success rate experienced at a forwarding peer has only increased from 3.5% to 6.9%. Over 9% of queries forwarded by a peer do not result in any query hits. With an average query size of over 1 bytes and 3 neighbors for an ultra-peer, this results in almost 1 GB of wasted bandwidth in a 24 hour session. We outline solution approaches to solve this problem and make P2P systems viable for a diverse range of applications. 1 Introduction In recent years, Peer-to-Peer (P2P) systems have become popular platforms for distributed and decentralized applications. P2P applications such as Gnutella[1], Kazaa [5] and Overnet/eDonkey [6] are widely used and as such, their behavior and performance have been studied widely [11, 3, 8]. Gnutella, although popular, has been shown to suffer from problems such as free-riding users, high bandwidth utilization and poor search performance. With the number of users in the network growing and the ultra-peer architecture becoming more dominant [9], solving the problems of bandwidth utilization and search performance becomes increasingly important to the long-term viability of Gnutella as well as other filesharing P2P applications. In this paper we present an analysis of the trends in Gnutella traffic 23, 25, and 26. First, we examined large-scale macro behaviors to determine how This work was supported in part by the U.S. National Science Foundation (IIS and CNS ).

2 Gnutella traffic evolves over the long term. We captured traffic in 23, 25, and 26 and compare the message rate, bandwidth utilization, and queueing properties of each Gnutella message type to determine the changes in characteristic behavior of Gnutella traffic over the long term. We found that the percentage of the bandwidth consumed by each message type changed considerably. We also studied more localized behavior such as the query success rate experienced by peers forwarding queries into the network. We found that the overwhelming majority (> 9%) of the queries forwarded by a peer do not yield any query hits back to the forwarding peer. We show that earlier changes to the Gnutella protocol had not achieved the intended benefits of alleviating poor search performance and high bandwidth utilization. 2 Gnutella Overview First we describe the Gnutella system. Gnutella is a popular P2P filesharing application. Gnutella users connect to each other to form an overlay network. This overlay is used to search for content shared by other peers. The Gnutella protocol is a distributed and decentralized protocol. Peers issue request messages that are flooded to all of their neighbors. Each neighboring peer, in turn, forwards the request to all of its neighbors until a specified time-to-live (TTL) is reached. This flooding mechanism allows for reaching many peers, but can quickly overwhelm the available network bandwidth. Next we describe the two major versions of the Gnutella protocol. 2.1 Protocol v.4 The original Gnutella protocol, v.4, assumed that each Gnutella peer was equal in terms of capacity and participation. The v.4 protocol specified four major messages: Ping, Pong, Query, and QueryHit. The protocol also specified other messages such as Push requests, but these messages were rare and did not represent a significant percentage of the messages sent in the Gnutella network. The Ping and Pong control messages are used by Gnutella clients to discover other peers in the network. Ping and Query messages are flooded to each neighbor. The flooding continues until the TTL for the request has been reached. Responses such as Pong and QueryHit messages are routed back to the originator along the path of the request. When a peer receives a Query message it evaluates the query string and determines if any of its shared content can satisfy the query. If so, the peer sends back a QueryHit message along the path the it received the Query. Each QueryHit message contains the following information: the number of objects that match the query, the IP address and port of the peer where the objects are located, and a set of results including the file name size of each matching object.

3 2.2 Protocol v.6 As Gnutella became popular, both the size of the network and the amount of traffic on the network increased. The increase in network traffic coupled with the poor Internet connections of many users [11] overwhelmed many peers in terms of routing and processing messages. In order to overcome the performance limitations of the original protocol, the v.6 [2] was introduced. The new v.6 ultrapeer architecture reduced the incoming query message rate thus lowering the bandwidth requirements to process queries. However, as we show in this paper, the architecture change did not eliminate the problem of high bandwidth consumption; it shifted it to a different place in the network. Specifically, the v.6 protocol requires the ultrapeers to maintain many connections. 3 Related Work Saroiu et. al. [11] and Ripeanu et. al. [1] studied several aspects of the Gnutella file-sharing network. The authors discovered that new clients tended to connect to Gnutella peers with many connections. This led to an overlay network whose graph representation had a node degree distribution of a power law graph. The authors also identified that most users do not share content (free-ride) and that a large percentage of nodes had poor network connectivity. More recently, the problem of free-riding on Gnutella was revisited in [4]. The authors found that a greater number of users, as a percentage of the total users, are free-riding on Gnutella than in 2. Queries forwarded to free-riding users will not yield any responses. Thus free-riding users degrade the high-level utility of the system and the low-level performance A more closely related work to this paper is by Rasti et al. [9]. Their aim was to analyze the evolution of the v.6 two-tier architecture with respect to the topological properties of the network. The authors measured an increase in the number of nodes in the network and showed that as nodes began to transition to the v.6 protocol, the network became unbalanced with respect to the number of leaf and ultra-peers. They showed that modifications to major Gnutella client software and rapid adoption rate by users helped restore the overlay s desired properties. Our work is similar in that we analyze Gnutella s evolution from the v.4 to the v.6 architecture. However, our work focuses on the evolution of traffic characteristics and query performance. This allows for better understanding of the effects of architectural changes to the system. Further, understanding the evolution of the system s workload enables better decisions about future changes to the system and facilitates better modeling and simulation of large-scale P2P networks. 4 Results 4.1 Methodology We captured traces for two weeks in May and June of 23 as well as October of 25, and June through September of 26. We modified the source code of Phex

4 [7], an open source Gnutella client written in Java to log traffic on the network. The client would connect to the network and log every incoming and outgoing message that passed through it. It is important to note that our traffic capturing client does not actively probe the system; it only logs messages that it receives and forwards. Each log entry contained the following fields: timestamp, Gnutella message ID, message direction (incoming/outgoing), message type, TTL, hops taken, and the size of the message. In addition to logging incoming and outgoing messages, the client also logged queries and query responses in separate log files. For queries, we logged the query string. In the case of query responses, the log contained the Gnutella message ID of the original query, the peer ID where the matching objects are located and a list of the matching objects which includes the filename and the size of the file. The original protocol did not have separate classes of peers. As such, the traces from 23 were collected with our client running as a regular peer. With the introduction of the v.6 protocol, peers were classified as either leaf or ultra-peers. Our traces from 25 and 26 were captured with our client acting as an ultra-peer. The client was allowed to run 24 hours a day and logs were partitioned into two hour interval for ease of processing. These traces give insight as to the evolution of the Gnutella traffic over the last several years. We use the data from these traces to evaluate our design choices. In the interest of space, we present only results from one 24 hour trace from a typical weekday. The same day of the week was used (Thursday) for the traces of each year. Analysis of other days show similar results as those presented in this paper with weekends showing only 3% more traffic than weekdays. 4.2 Summary of Data The average number of messages (incoming and outgoing) handled by the client was 2.5M in 23 and 2.67M in 26 per 2 hour interval. The average file size for a 2 hour trace was 292MB in 23 and 253MB in 26. This represents over 3GB of data per 24 hour period. In 23, a 24 hour trace saw 2,784 different peers directly connected to it. In 26, our client only saw 1,155 different peers directly connected to it in a 24 period. Although our 26 traces saw fewer peers, other studies [8, 9] show that the network has grown over time. The reduction in number of peers seen by our client can be attributed to a trend toward longer session times. In 23, more than 95% of the sessions lasted less than 3 minutes. In 26, more than 5% of the sessions lasted longer than 6 minutes. Longer session times imply slower churn rate for the connections at each node and thus our client would not see as many connections when the session times for peers is longer. It is important to note that our data from 25 shows characteristics that are consistent with the evolution of the protocol with respect to TTL and hops taken for messages. However, the bandwidth data is somewhat skewed. Investigation of this data revealed that although the client attempted to act as an ultra-peer, it was not able to maintain many connections (< 1). In contrast, the same client in 26 was able to maintain over 3 connections consistently. Our traces from a

5 client running as a leaf node show different behavior with a significant reduction in bandwidth for leaf nodes in Message TTL Analysis Table 1. Mean TTL left and Hops Taken for different message types in a 24 hour period from 23, 25, and 26 data All Messages Control Messages Query Messages TTL Left Hops Taken TTL Left Hops Taken TTL Left Hops Taken The TTL and the number of hops taken for each give an indication of how far messages travel in the network. Table 1 shows the mean TTL and number of hops taken for different classes of messages in a 24 hour trace for 23, 25 and 26 traffic. In 23, messages travel 6.85 hops into the network. Messages required 4.46 hops to reach our logging client and arrived with a mean TTL left of Similarly, in 25, messages travel 6.31 hops into the network requiring 3.41 hops to reach our client and arriving with a TTL of In contrast, traffic from 26 travels 3.86 hops into the network requiring 3.34 hops to reach our client and arriving with a mean TTL left of.52. This change can be attributed to the v.6 protocol. The new protocol typically restricts the initial TTL of the message to between 4 and 5 hops to reduce the burden on the network due flooding from ultra-peers with many neighbors. A closer inspection of the TTL left and hops taken for the different types of messages reveals a change in the traffic characteristics of the different message types. In 23, control messages (Ping and Pong) traveled further into the network and had to be propagated to many peers. On average, 23 control traffic had a mean of 3.1 hops taken and a mean TTL left of 2.8. In contrast, 26 control traffic had a mean of 3.4 hops taken and a mean TTL left of.37. Although messages arrive at a node from similar distances in 23 and 26, they are not expected to be propagated as far in 26 as they were in 23. Unlike control traffic, query traffic in 26 is expected to be propagated further at each node. In addition, the TTL left for query traffic has remained fairly stable from 23 to 26: 2.36 in 23 and 2.26 in 26. We should note the trend between 23, 25 and 26. In 25, control messages traveled deep into the network (6.93 hops) compared to 26 (3.79 hops). In contrast, query messages traveled 6.93 hops in 23, 5.23 hops in 25 and 4.68 in 26. This shows that query messages did not travel as deep into the network in 25 as they did in 23. This can be attributed to the transitioning to the current state of the v.6 protocol. The protocol specification only sets

6 guidelines as to what values to set for the initial TTL of a message. In 25, vendors of Gnutella clients had already begun to set lower TTL values for query messages, but had continued to use a TTL of 7 for control messages. In 26, these vendors transitioned to lower TTL values for all messages. In Figures 1 and 2 we show the hourly mean TTL and hops taken for control and query traffic respectively. The traffic represents data from 24 hour traces from 23 and 26. We note that the TTL left and hops taken for each message remain relatively constant throughout the day even though bandwidth consumption and the number of nodes in the network fluctuate throughout the day. This observation can be used to develop traffic models when simulating a P2P filesharing workload Traffic 7 23 Traffic 6 26 Traffic 6 26 Traffic 5 5 TTL 4 3 Hops Taken (a) TTL left (b) Hops Taken Fig. 1. Mean hourly TTL and hops taken for incoming control messages in a 24 hour period for 23 and 26 traffic. 4.4 Bandwidth Analysis In this section we analyze the traces to identify changes in bandwidth consumption. We track the bandwidth utilization for control (Ping and Pong), query (queries and query hits), and route table update messages. We limit the control messages only to the basic Ping and Pong messages as they make up the overwhelming majority of non-query messages. Additionally, by tracking route table updates separate from the basic control messages, we can determine how the behavior of the network has evolved with respect to the use of each message type. Note that we do not show bandwidth consumed by peers for downloading objects. We examined bandwidth utilization by tracking the bandwidth requirements for each message type. Figure 3 shows the bandwidth utilization for each message type in a 24 hour period for 23 and 26 traffic respectively. In 23, query traffic dominates the bandwidth utilization. Further, the amount of query traffic

7 7 23 Traffic 7 23 Traffic 6 26 Traffic 6 26 Traffic 5 5 TTL 4 3 Hops Taken (a) TTL left (b) Hops Taken Fig. 2. Mean hourly TTL and hops taken for incoming query messages in a 24 hour period for 23 and 26 traffic. varies at different times of the day while control traffic remains relatively stable throughout the entire day. The 26 traffic exhibits different properties. First, the query traffic represents significantly less bandwidth than in 23, although out going query traffic is still the dominant consumer of bandwidth. Control traffic, both incoming and outgoing, plays a larger role under the current v.6 protocol in 26 than it did in 23, as does route table update traffic. The changes in bandwidth throughout the day are most pronounced in incoming and outgoing query traffic while control traffic remains relatively constant throughout the day. The increase in route table update traffic can be attributed to the fact that in 23, very few clients were using the new v.6 protocol. As more clients moved to the new protocol, we see a rise in the number of route table update messages sent in the system. The evolution of the Gnutella protocol and network results in traffic characteristics that are different in 26 than they were in 23. First, in 23, control and route table updates represent a small percentage of the total bandwidth utilization. In 26, however, control traffic is significantly more prominent with respect to the percentage of bandwidth utilization. Additionally, route table updates become a non-trivial message category in 26 whereas these messages were virtually non-existent in 23. Finally, the characteristics of query traffic from 23 to 26 change dramatically. In 23, query traffic (both incoming and outgoing) dominates the bandwidth utilization. In 26, query traffic still consumes a large percentage of the bandwidth. However, only outgoing query traffic consumes a large amount of bandwidth. Incoming query traffic uses less bandwidth than control traffic in 26. This is a result of the evolution of the protocol specification. The v.6 protocol establishes ultra-peers as peers with many neighbors. Ultra-peers can collect information about files shared at their leaf nodes and use this information to respond to queries. We saw earlier that message TTL are decreased in the v.6 protocol, so fewer messages are sent in the network. Additionally, the v.6 query routing protocol attempts to minimize

8 25 2 In Control Traffic Out Control Traffic In Query Traffic Out Query Traffic In Other Traffic Out Other Traffic 25 2 In Control Traffic Out Control Traffic In Query Traffic Out Query Traffic In Other Traffic Out Other Traffic Bandwidth (kbps) 15 1 Bandwidth (kbps) (a) 23 Traffic (b) 26 Traffic Fig. 3. Hourly bandwidth for different message types in a 24 hour period for 23 and 26 traffic. Note that the curves in Figure 3(a) do not wrap around to the same value at the end of a 24 hour period. This was due to our gathering procedure in 23 that required the client to shut down after each 2 hour trace for processing. The client did not get restarted again for another hour in order to accommodate processing. This problem was fixed in our later data capturing process. the number of query messages that are sent. Therefore, a node under the current v.6 protocol will receive fewer incoming query messages than it would have in 23. Because ultra-peers have many connections, propagating queries to each neighbor results in outgoing query traffic utilizing a large percentage of the total bandwidth. Although the outgoing query bandwidth is large in absolute terms, there is an improvement relative to 23 since the outgoing query bandwidth was reduced from 2 kbps to 1 kbps when the node is an ultra-peer with 3 connections. Leaf peers benefit from this situation as they have fewer connections and receive very few incoming queries. 4.5 Query Traffic Analysis In the previous section we showed the evolution of bandwidth utilization for the different message types from 23 to 26. In this section, we will show the evolution of query traffic. We are interested in examining the relationship between the number of query messages sent and received and the success rate. Table 2 shows a summary of the query traffic measurements for traces from 23 and 26 traffic. The values in the table are for a typical 24 hour interval. In 23, query traffic constituted a large component of the bandwidth used. We see in the table that in 23, a peer would receive over 5M query messages in a 24 hour interval, or approximately 6 queries per second. In 26, this number is significantly reduced to 28K queries in a 24 hour interval, or about 3 queries per second. With a mean query message size of bytes for 23 traffic, 6 queries per second corresponds to an incoming data rate of 4.39 KB/s or 35.2 kbps. Such a data rate would overwhelm most home dial-up connections. The

9 Table 2. Query traffic summary for a 24 hour trace from 23, 25 and Queries Received 5,261,64 614, ,235 Mean Queries per second Query Message Size (including header) Successful Queries 184,14 63,69 19,443 Success Rate 3.5% 1.3% 6.9% Mean Queue Time (successful queries) s s s Mean Outgoing Messages per Query Mean Outgoing Bytes per Query.285 kb.677kb 4.26 kb Mean Outgoing Query Bandwidth kbps 38.5 Kbps 14.3 kbps current state of the network has reduced the number of queries received at each node at the expense of requiring a large number of connections for ultra-peer nodes. The large number of connections, in turn, results in a large outgoing bandwidth utilization. On a per query basis, 23 traffic generated.285 KB per query, while 26 traffic generated 4.26 KB per query. In 23, each query was propagated to less than 4 peers on average while in 26, each query was propagated to a mean of 38 peers. Note that in 25, the bandwidth utilization is much lower compared to both 23 and 26. This is because our client was not able to operate as an ultra-peer and thus only was able to maintain less than 1 connections. However, other metrics such as incoming query rate for 25 is consistent with the trend from 23 to 26. Next we investigate the ability of the network to successfully resolve queries. We see that from 23 to 26, the success rate almost doubles from 3.5% in 23 to 6.9% in 26. Nevertheless, the success rate is still remarkably low. In 26, a success rate of 6.9% implies that 93.1% of queries that reach a node will not be resolved after the node propagates the query. This means that each node is utilizing bandwidth to process queries but that effort is wasted on over 9% of those queries. In a 24 hour interval, a node receives approximately 28K queries with an average size of 15 bytes. Each of these queries is propagated to approximately 38 neighbors resulting in 1.14 GB of outgoing data, of which 93.1%, or 1.3 GB, are wasted since the success rate is only 6.9%. 5 Limitations The study described in this paper was conducted using a single ultrapeer on our university s campus. As such, it may not be representative of the global behavior of Gnutella. We had performed measurements from a peer on a broadband network using a different service provider and observed similar results to as was reported in this paper. Also, the broadband studies were conducted in 26; we do not have any measurement from 23 or 25 for this analysis. Hence, we cannot ascertain the consistency of the results among broadband users

10 6 Conclusion We presented the results of our measurement and analysis of the Gnutella filesharing network. We showed that although the Gnutella architecture changed from 23 to 26 to help alleviate query bandwidth utilization, the success rate of queries has not shown significant improvements. Additionally, the bandwidth utilization problem is alleviated at low-capacity peers (leaf peers), but high capacity peers (ultra peers) experience an increase in query bandwidth utilization. The increase in bandwidth occurs due to the large number of neighbors connected to the ultra peers. These findings indicate that despite the efforts to improve the performance of Gnutella, search performance is limited by the design of the protocol. We are investigating an approach that exploits the underlying characteristics of the queries and the distribution of objects in the system in order to improve search performance. References 1. Gnutella protocol v Gnutella protocol v draft.html. 3. Krishna P. Gummadi, Richard J. Dunn, Stefan Saroiu, Steven D. Gribble, Henry M. Levy, and John Zahorjan. Measurement, modeling, and analysis of a peer-to-peer file-sharing workload. In Proceedings of the nineteenth ACM symposium on Operating systems principles, pages ACM Press, Daniel Hughes, Geoff Coulson, and James Walkerdine. Free riding on gnutella revisited: The bell tolls? IEEE Distributed Systems Online, 6(6), June Kazaa media desktop. 6. Overnet. 7. The phex gnutella client. 8. Yi Qiao and Fabin E. Bustamante. Structured and unstructured overlays under the microscope - a measurement-based view of two p2p systems that people use. In Proceedings of the USENIX Annual Technical Conference, Amir H. Rasti, Daniel Stutzbach, and Reza Rejaie. On the long-term evolution of the two-tier gnutella overlay. In IEEE Golbal Internet, M. Ripeanu, I. Foster, and A. Iamnitchi. Mapping the gnutella network: Properties of large-scale peer-to-peer systems and implications for system design. IEEE Internet Computing Journal, 6(1), Stefan Saroiu, P. Krishna Gummadi, and Steven D. Gribble. A measurement study of peer-to-peer file sharing systems. In Proceedings of Multimedia Computing and Networking 22 (MMCN 2), San Jose, CA, USA, January 22.

On the Penetration of Business Networks by P2P File Sharing

On the Penetration of Business Networks by P2P File Sharing On the Penetration of Business Networks by P2P File Sharing Kevin Lee School of Computer Science, University of Manchester, Manchester, UK. +44 () 161 2756132 klee@cs.man.ac.uk Danny Hughes Computing,

More information

A Measurement Study of Peer-to-Peer File Sharing Systems

A Measurement Study of Peer-to-Peer File Sharing Systems CSF641 P2P Computing 點 對 點 計 算 A Measurement Study of Peer-to-Peer File Sharing Systems Stefan Saroiu, P. Krishna Gummadi, and Steven D. Gribble Department of Computer Science and Engineering University

More information

An Open Tracing System for P2P File Sharing Systems

An Open Tracing System for P2P File Sharing Systems An Open Tracing System for P2P File Sharing Systems Danny Hughes, James Walkerdine Computing Department Lancaster University Lancaster, UK danny@comp.lancs.ac.uk walkerdi@comp.lancs.ac.uk Kevin Lee School

More information

On the Penetration of Business Networks by P2P File Sharing

On the Penetration of Business Networks by P2P File Sharing On the Penetration of Business Networks by P2P File Sharing Kevin Lee School of Computer Science, University of Manchester, Manchester, M13 9PL, UK. +44 (0) 161 2756132 klee@cs.man.ac.uk Danny Hughes Computing,

More information

Topology Awareness in Unstructured P2P Networks using Accessible Bandwidth

Topology Awareness in Unstructured P2P Networks using Accessible Bandwidth Topology Awareness in Unstructured P2P Networks using Accessible Bandwidth B. Lalitha Department of CSE JNTUCEA, Anantapur, India Ch D V Subba Rao, Ph.D Department of CSE S V University, Tirupathi, India

More information

Gnutella-Pro: What bandwidth barrier?

Gnutella-Pro: What bandwidth barrier? Gnutella-Pro: What bandwidth barrier? Richard Massey, Shriram Bharath & Ankur Jain {rmassey,sbharath,ajain}@cs.ucsd.edu 1. Introduction This paper discusses the history, and evolution of the peer-to-peer

More information

Simulating a P2P File-Sharing Network

Simulating a P2P File-Sharing Network Simulating a P2P File-Sharing Network 1 Mario T. Schlosser and Sepandar D. Kamvar Abstract Assessing the performance of peer-to-peer algorithms such as topology construction protocols, reputation algorithms

More information

Mapping the Gnutella Network. What are P2P Networks?

Mapping the Gnutella Network. What are P2P Networks? Mapping the Gnutella Network Matei Ripeanu (Univ. of Chicago) Adriana Iamnitchi (Univ. of Chicago) Ian Foster (Univ. of Chicago & Argonne National Laboratory) What are P2P Networks? Networks to share information

More information

to-peer Networking: Contents Mapping the Gnutella Network Chinhua Lin

to-peer Networking: Contents Mapping the Gnutella Network Chinhua Lin Peer-to to-peer Networking: Mapping the Gnutella Network Chinhua Lin 11/5/2002 Mapping the Gnutella 1 Contents Why Peer-to-Peer? Categories of P2P systems Gnutella Protocol Network Analysis Potential Improvements

More information

Study and Bridging of Peer-to-Peer File Sharing Systems

Study and Bridging of Peer-to-Peer File Sharing Systems Study and Bridging of Peer-to-Peer File Sharing Systems Georgios Portokalidis 1, Evangelos P. Markatos 2 Manolis Marazakis Institute of Computer Science, Foundation for Research and Technology Hellas PO

More information

Moderating Redundant Traffic Generated by Topology Mismatch in P2P Networks

Moderating Redundant Traffic Generated by Topology Mismatch in P2P Networks Moderating Redundant Traffic Generated by Topology Mismatch in P2P Networks B. Lalitha Ch D V Subba Rao, Ph.D Department of CSE JNTUCEA, Anantapur, India Department of CSE S V University, Tirupathi, India

More information

Analysis of the Traffic on the Gnutella Network

Analysis of the Traffic on the Gnutella Network Analysis of the Traffic on the Gnutella Network Kelsey Anderson University of California, San Diego CSE222 Final Project March 21 Abstract The Gnutella network is an overlay network

More information

Simulating a File-Sharing P2P Network

Simulating a File-Sharing P2P Network Simulating a File-Sharing P2P Network Mario T. Schlosser, Tyson E. Condie, and Sepandar D. Kamvar Department of Computer Science Stanford University, Stanford, CA 94305, USA Abstract. Assessing the performance

More information

The Case for a Hybrid P2P Search Infrastructure

The Case for a Hybrid P2P Search Infrastructure The Case for a Hybrid P2P Search Infrastructure Boon Thau Loo Ryan Huebsch Ion Stoica Joseph M. Hellerstein University of California at Berkeley Intel Research Berkeley boonloo, huebsch, istoica, jmh @cs.berkeley.edu

More information

The Role and uses of Peer-to-Peer in file-sharing. Computer Communication & Distributed Systems EDA 390

The Role and uses of Peer-to-Peer in file-sharing. Computer Communication & Distributed Systems EDA 390 The Role and uses of Peer-to-Peer in file-sharing Computer Communication & Distributed Systems EDA 390 Jenny Bengtsson Prarthanaa Khokar jenben@dtek.chalmers.se prarthan@dtek.chalmers.se Gothenburg, May

More information

An Analysis of Internet Content Delivery Systems (and more!)

An Analysis of Internet Content Delivery Systems (and more!) An Analysis of Internet Content Delivery Systems (and more!) Stefan Saroiu, Krishna Gummadi, Richard J.Dunn, Steven D. Gribble, and Henry M. Levy Proceedings of the 5th Symposium on Operating Systems Design

More information

Monitoring Challenges and Approaches for P2P File-Sharing Systems

Monitoring Challenges and Approaches for P2P File-Sharing Systems Monitoring Challenges and Approaches for P2P File-Sharing Systems Danny Hughes Computing, InfoLab21, Lancaster University, Lancaster, UK. +44 (0)1524 510351 danny@comp.lancs.ac.uk James Walkerdine Computing,

More information

Characterizing the Query Behavior in Peer-to-Peer File Sharing Systems*

Characterizing the Query Behavior in Peer-to-Peer File Sharing Systems* Characterizing the Query Behavior in Peer-to-Peer File Sharing Systems* Alexander Klemm a Christoph Lindemann a Mary K. Vernon b Oliver P. Waldhorst a ABSTRACT This paper characterizes the query behavior

More information

Unstructured Networks: Search. Márk Jelasity

Unstructured Networks: Search. Márk Jelasity Unstructured Networks: Search Márk Jelasity Outline Emergence of decentralized networks The Gnutella network: how it worked and looked like Search in unstructured networks Random walk search in power law

More information

On the Equivalence of Forward and Reverse Query Caching in Peer-to-Peer Overlay Networks

On the Equivalence of Forward and Reverse Query Caching in Peer-to-Peer Overlay Networks On the Equivalence of Forward and Reverse Query Caching in Peer-to-Peer Overlay Networks Ali Raza Butt 1, Nipoon Malhotra 1, Sunil Patro 2, and Y. Charlie Hu 1 1 Purdue University, West Lafayette IN 47907,

More information

International Journal of Scientific & Engineering Research, Volume 4, Issue 11, November-2013 349 ISSN 2229-5518

International Journal of Scientific & Engineering Research, Volume 4, Issue 11, November-2013 349 ISSN 2229-5518 International Journal of Scientific & Engineering Research, Volume 4, Issue 11, November-2013 349 Load Balancing Heterogeneous Request in DHT-based P2P Systems Mrs. Yogita A. Dalvi Dr. R. Shankar Mr. Atesh

More information

Counteracting free riding in Peer-to-Peer networks q

Counteracting free riding in Peer-to-Peer networks q Available online at www.sciencedirect.com Computer Networks 52 (2008) 675 694 www.elsevier.com/locate/comnet Counteracting free riding in Peer-to-Peer networks q Murat Karakaya *, _ Ibrahim Körpeoğlu,

More information

Kent State University. Peer-to-Peer Systems & Applications. A Course on Foundations of CS 6/ Applications & Systems. Foundation of Peer-to-Peer

Kent State University. Peer-to-Peer Systems & Applications. A Course on Foundations of CS 6/ Applications & Systems. Foundation of Peer-to-Peer A Course on Foundations of Peer-to-Peer Systems & Applications LECT-07, S-1 CS 6/75995 Foundation of Peer-to-Peer Applications & Systems Kent State University Dept. of Computer Science www.cs.kent.edu/~javed/class-p2p08/

More information

2.6.2 Searching for Information in a P2P Community

2.6.2 Searching for Information in a P2P Community 2.6 PEER-TO-PEER APPLICATIONS 5 are satisfied with the trading, they will put each other in their top four lists and continue trading with each other until one of the peers finds a better partner. The

More information

Understanding the Roles of Servers in Large-scale Peer-Assisted Online Storage Systems

Understanding the Roles of Servers in Large-scale Peer-Assisted Online Storage Systems Understanding the Roles of Servers in Large-scale Peer-Assisted Online Storage Systems Fangming Liu, Ye Sun, Bo Li, Xinyan Zhang Hong Kong University of Science & Technology, Roxbeam Inc. Abstract Online

More information

Abstract. 1. Introduction

Abstract. 1. Introduction Supporting s in Gnutella File Sharing Network with Agents Tim Hsin-ting Hu, Binh Thai, Aruna Seneviratne School of Electrical Engineering and Telecommunications University of New South Wales, Sydney, Australia

More information

International journal of Engineering Research-Online A Peer Reviewed International Journal Articles available online http://www.ijoer.

International journal of Engineering Research-Online A Peer Reviewed International Journal Articles available online http://www.ijoer. RESEARCH ARTICLE ISSN: 2321-7758 GLOBAL LOAD DISTRIBUTION USING SKIP GRAPH, BATON AND CHORD J.K.JEEVITHA, B.KARTHIKA* Information Technology,PSNA College of Engineering & Technology, Dindigul, India Article

More information

Clustering in Peer-to-Peer File Sharing Workloads

Clustering in Peer-to-Peer File Sharing Workloads Clustering in Peer-to-Peer File Sharing Workloads F. Le Fessant, S. Handurukande, A.-M. Kermarrec & L. Massoulié INRIA-Futurs and LIX, Palaiseau, France Distributed Programming Laboratory, EPFL, Switzerland

More information

Decentralized Peer-to-Peer Network Architecture: Gnutella and Freenet

Decentralized Peer-to-Peer Network Architecture: Gnutella and Freenet Decentralized Peer-to-Peer Network Architecture: Gnutella and Freenet AUTHOR: Jem E. Berkes umberkes@cc.umanitoba.ca University of Manitoba Winnipeg, Manitoba Canada April 9, 2003 Introduction Although

More information

Using UDP Packets to Detect P2P File Sharing

Using UDP Packets to Detect P2P File Sharing 188 IJCSNS International Journal of Computer Science and Network Security, VOL.7 No.8, August 27 Using UDP Packets to Detect P2P File Sharing Tsang-Long Pao and Jian-Bo Chen Tatung University, Taipei,

More information

Capturing Accurate Snapshots of the Gnutella Network

Capturing Accurate Snapshots of the Gnutella Network Capturing Accurate Snapshots of the Gnutella Network Daniel Stutzbach, Reza Rejaie Department of Computer and Information Science University of Oregon {agthorr,reza}@cs.uoregon.edu Abstract A common approach

More information

Mapping the Gnutella Network: Macroscopic Properties of Large-Scale Peer-to-Peer Systems

Mapping the Gnutella Network: Macroscopic Properties of Large-Scale Peer-to-Peer Systems Mapping the Gnutella Network: Macroscopic Properties of Large-Scale Peer-to-Peer Systems Matei Ripeanu, Ian Foster {matei, foster}@cs.uchicago.edu Abstract Despite recent excitement generated by the peer-to-peer

More information

P2P support for web proxy caching with web characteristics

P2P support for web proxy caching with web characteristics P2P support for web proxy caching with web characteristics Kyungbaek Kim 1 and Daeyeon Park 1 Department of Electrical Engineering & Computer Science, Division of Electrical Engineering, Korea Advanced

More information

Towards a Better Understanding of Churn in Peer-to-Peer Networks

Towards a Better Understanding of Churn in Peer-to-Peer Networks TECHNICAL REPORT CIS-TR-04-06, DEPARTMENT OF COMPUTER SCIENCE, UNIVERSITY OF OREGON, NOVEMBER 2004 1 Towards a Better Understanding of Churn in Peer-to-Peer Networks Daniel Stutzbach, Reza Rejaie agthorr,

More information

RESEARCH ISSUES IN PEER-TO-PEER DATA MANAGEMENT

RESEARCH ISSUES IN PEER-TO-PEER DATA MANAGEMENT RESEARCH ISSUES IN PEER-TO-PEER DATA MANAGEMENT Bilkent University 1 OUTLINE P2P computing systems Representative P2P systems P2P data management Incentive mechanisms Concluding remarks Bilkent University

More information

The edonkey File-Sharing Network

The edonkey File-Sharing Network The edonkey File-Sharing Network Oliver Heckmann, Axel Bock, Andreas Mauthe, Ralf Steinmetz Multimedia Kommunikation (KOM) Technische Universität Darmstadt Merckstr. 25, 64293 Darmstadt (heckmann, bock,

More information

Optimizing and Balancing Load in Fully Distributed P2P File Sharing Systems

Optimizing and Balancing Load in Fully Distributed P2P File Sharing Systems Optimizing and Balancing Load in Fully Distributed P2P File Sharing Systems (Scalable and Efficient Keyword Searching) Anh-Tuan Gai INRIA Rocquencourt anh-tuan.gai@inria.fr Laurent Viennot INRIA Rocquencourt

More information

Real Datasets for File-sharing Peer-to-Peer Systems

Real Datasets for File-sharing Peer-to-Peer Systems Real Datasets for File-sharing Peer-to-Peer Systems Shen Tat Goh 1, Panos Kalnis 1, Spiridon Bakiras 2, and Kian-Lee Tan 1 1 Department of Computer Science National University of Singapore 3 Science Drive

More information

Building a Scalable Bipartite P2P Overlay Network

Building a Scalable Bipartite P2P Overlay Network Building a Scalable Bipartite P2P Overlay Network Yunhao Liu, Li Xiao Dept. of Computer Science and Engineering Michigan State University East Lansing, MI 48824, USA {liuyunha, lxiao}@cse.msu.edu Lionel

More information

LECT-07, S-1 FP2P08S, Javed I.

LECT-07, S-1 FP2P08S, Javed I. A Course on Foundations of Peer-to-Peer Systems & Applications LECT-07, S-1 CS 6/75995 Foundation of Peer-to-Peer Applications & Systems Kent State University Dept. of Computer Science www.cs.kent.edu/~javed/class-p2p08/

More information

CS 360 Internet Programming

CS 360 Internet Programming Peer-to-Peer Networking File Sharing, and Computer Science Brigham Young University 1 Peer-to-Peer File Sharing with Ultrapeers Overview Peer-to-Peer File Sharing with Ultrapeers peer-to-peer peers exchange

More information

Peer-to-Peer Networks 02: Napster & Gnutella. Christian Schindelhauer Technical Faculty Computer-Networks and Telematics University of Freiburg

Peer-to-Peer Networks 02: Napster & Gnutella. Christian Schindelhauer Technical Faculty Computer-Networks and Telematics University of Freiburg Peer-to-Peer Networks 02: Napster & Gnutella Christian Schindelhauer Technical Faculty Computer-Networks and Telematics University of Freiburg Napster Shawn (Napster) Fanning - published 1999 his beta

More information

P2P networks, free riding, and counter-mechanisms

P2P networks, free riding, and counter-mechanisms P2P networks,, and counter-mechanisms İbrahim Körpeoğlu Department Computer Engineering Bilkent University Seminar Place: Department Electrical and Electronics Engineering Bilkent University, Ankara Date:

More information

Security Problems in P2P Networks: A Case Study on the Gnutella Network

Security Problems in P2P Networks: A Case Study on the Gnutella Network Security Problems in P2P Networks: A Case Study on the Gnutella Network Wei Wang, Dragos Andrei and Huan Song Abstract Peer-to-peer (P2P) file-sharing networks are becoming more and more popular recently

More information

The Case for a Hybrid P2P Search Infrastructure

The Case for a Hybrid P2P Search Infrastructure University of Pennsylvania ScholarlyCommons Departmental Papers (CIS) Department of Computer & Information Science February 2004 The Case for a Hybrid P2P Search Infrastructure Boon Thau Loo University

More information

CS 580 Client-Server Programming Spring Semester, 2004 Doc 20 Gnutella Contents

CS 580 Client-Server Programming Spring Semester, 2004 Doc 20 Gnutella Contents 3/23/04 Doc 20 Gnutella slide # 1 CS 580 Client-Server Programming Spring Semester, 2004 Doc 20 Gnutella Contents Reference... 1 Gnutella... 2 Requests and Responses... 5 Header... 6 Ping 0x00... 8 Pong

More information

Peer-to-Peer File Sharing

Peer-to-Peer File Sharing The Effects of File Sharing on a Service Provider's Network An Industry White Paper Copyright July 2002, Sandvine Incorporated www.sandvine.com 408 Albert Street Waterloo, Ontario Canada N2L 3V3 Executive

More information

Application Layer Traffic Analysis of a Peer-to-Peer System

Application Layer Traffic Analysis of a Peer-to-Peer System Application Layer Traffic Analysis of a Peer-to-Peer System Dietmar Tutsch Technical University Berlin Institute of Computer Engineering and Microelectronics Berlin, Germany DietmarT@cs.tu-berlin.de Gilbert

More information

Clustering in Peer-to-Peer File Sharing Workloads

Clustering in Peer-to-Peer File Sharing Workloads Clustering in Peer-to-Peer File Sharing Workloads F. Le Fessant, S. Handurukande, A.-M. Kermarrec & L. Massoulié INRIA-Futurs and LIX, Palaiseau, France EPFL, Lausanne, Switzerland Microsoft Research,

More information

Efficient processing of rare queries Gnutella using a hybrid infrastructure

Efficient processing of rare queries Gnutella using a hybrid infrastructure Efficient processing of rare queries Gnutella using a hybrid infrastructure Mark Kornfilt½and Manfred Hauswirth¾ ½LimeWire LLC ¾DERI Galway Abstract. Gnutella is still one of the most popular P2P systems

More information

PEER TO PEER FILE SHARING USING NETWORK CODING

PEER TO PEER FILE SHARING USING NETWORK CODING PEER TO PEER FILE SHARING USING NETWORK CODING Ajay Choudhary 1, Nilesh Akhade 2, Aditya Narke 3, Ajit Deshmane 4 Department of Computer Engineering, University of Pune Imperial College of Engineering

More information

Implementation of a P2P system within a network simulation framework

Implementation of a P2P system within a network simulation framework Implementation of a P2P system within a network simulation framework Vinay Aggarwal Anja Feldmann Sebastian Mohr Technische Universität München, Germany {vinay,anja,mohrs}@net.in.tum.de Abstract To enable

More information

Load Balancing in Structured Peer to Peer Systems

Load Balancing in Structured Peer to Peer Systems Load Balancing in Structured Peer to Peer Systems DR.K.P.KALIYAMURTHIE 1, D.PARAMESWARI 2 Professor and Head, Dept. of IT, Bharath University, Chennai-600 073 1 Asst. Prof. (SG), Dept. of Computer Applications,

More information

Load Balancing in Structured Peer to Peer Systems

Load Balancing in Structured Peer to Peer Systems Load Balancing in Structured Peer to Peer Systems Dr.K.P.Kaliyamurthie 1, D.Parameswari 2 1.Professor and Head, Dept. of IT, Bharath University, Chennai-600 073. 2.Asst. Prof.(SG), Dept. of Computer Applications,

More information

A reputation-based trust management in peer-to-peer network systems

A reputation-based trust management in peer-to-peer network systems A reputation-based trust management in peer-to-peer network systems Natalia Stakhanova, Sergio Ferrero, Johnny Wong, Ying Cai Department of Computer Science Iowa State University Ames, Iowa 50011 USA {

More information

N6Lookup( title ) Client

N6Lookup( title ) Client CS 640: Introduction Networks AdityaAkella Peer-to-Peer Lecture 24 -to Computer p2p Uses Downloading: Searching Centralized Flooding Smarter Routing file of sharing p2p The (Freenet, (Gnutella, flooding

More information

An Empirical Study of Free-Riding Behavior in the Maze P2P File-Sharing System

An Empirical Study of Free-Riding Behavior in the Maze P2P File-Sharing System An Empirical Study of Free-Riding Behavior in the Maze P2P File-Sharing System Mao Yang Beijing University ym@net.pku.edu.cn Zheng Zhang Microsoft Research Asia zzhang@microsoft.com Xiaoming Li Beijing

More information

2. Distributed Hash Tables (Chord)

2. Distributed Hash Tables (Chord) 2. Distributed Hash Tables (Chord) Hashing of search keys AND peer addresses on binary keys of length m e.g. m=8, key("jingle-bells.mp3")=17, key(196.178.0.1)=3 Data keys are stored at next larger node

More information

Multicast vs. P2P for Content Distribution

Multicast vs. P2P for Content Distribution Multicast vs. P2P for Content Distribution Subhashni Balakrishnan Department of Computer Science Rutgers University October 27 th 2008 Abstract: Content Distribution is originally an application developed

More information

Impact of Peer Incentives on the Dissemination of Polluted Content

Impact of Peer Incentives on the Dissemination of Polluted Content Impact of Peer Incentives on the Dissemination of Polluted Content Fabricio Benevenuto fabricio@dcc.ufmg.br Virgilio Almeida virgilio@dcc.ufmg.br Cristiano Costa krusty@dcc.ufmg.br Jussara Almeida jussara@dcc.ufmg.br

More information

Advanced Peer to Peer Discovery and Interaction Framework

Advanced Peer to Peer Discovery and Interaction Framework Advanced Peer to Peer Discovery and Interaction Framework Peeyush Tugnawat J.D. Edwards and Company One, Technology Way, Denver, CO 80237 peeyush_tugnawat@jdedwards.com Mohamed E. Fayad Computer Engineering

More information

Peer to Peer networks. some slides courtesy James

Peer to Peer networks. some slides courtesy James Peer to Peer networks some slides courtesy James Allan@umass 1 P2P example 2 P2P networks Alice wants the song foo she turns on Limewire Bob has the P2P application turned on and foo in the shared folder

More information

A methodology for workload characterization of file-sharing peer-to-peer networks

A methodology for workload characterization of file-sharing peer-to-peer networks A methodology for workload characterization of file-sharing peer-to-peer networks Diêgo Nogueira, Leonardo Rocha, Juliano Santos, Paulo Araújo, Virgílio Almeida, Wagner Meira Jr. Department of Computer

More information

Characterizing Unstructured Overlay Topologies in Modern P2P File-Sharing Systems

Characterizing Unstructured Overlay Topologies in Modern P2P File-Sharing Systems Characterizing Unstructured Overlay Topologies in Modern P2P File-Sharing Systems Daniel Stutzbach, Reza Rejaie University of Oregon {agthorr,reza}@cs.uoregon.edu Subhabrata Sen AT&T Labs Research sen@research.att.com

More information

Load Balancing through Automated Replication in Unstructured P2P File Sharing Systems

Load Balancing through Automated Replication in Unstructured P2P File Sharing Systems SBRC 2007 - Desempenho em Sistemas P2P 957 Load Balancing through Automated Replication in Unstructured P2P File Sharing Systems Cristina L. Abad 1 1 Facultad de Ingeniería en Electricidad y Computación

More information

THE INDEX POISONING ATTACK in P2P FILE SHARING SYSTEMS ABASTRACT

THE INDEX POISONING ATTACK in P2P FILE SHARING SYSTEMS ABASTRACT THE INDEX POISONING ATTACK in P2P FILE SHARING SYSTEMS ABASTRACT In P2P file-sharing users search index to find locations of desired files. Index poisoning, for particular file, attacks index with bogus

More information

The Gnutella Protocol Specification v0.4

The Gnutella Protocol Specification v0.4 The Gnutella Protocol Specification v0.4 Clip2 Distributed Search Services http://dss.clip2.com dss-protocols@clip2.com Gnutella (pronounced "newtella") is a protocol for distributed search. Although the

More information

Peer to peer networks: sharing between peers. Trond Aspelund

Peer to peer networks: sharing between peers. Trond Aspelund Peer to peer networks: sharing between peers Trond Aspelund Abstract In this literature survey we look at peer-to-peer networks. We first see how peer-to-peer networks distinguish themselves from the client/server

More information

Comparative Traffic Analysis Study of Popular Applications

Comparative Traffic Analysis Study of Popular Applications Comparative Traffic Analysis Study of Popular Applications Zoltán Móczár and Sándor Molnár High Speed Networks Laboratory Dept. of Telecommunications and Media Informatics Budapest Univ. of Technology

More information

Balanced Reputation Detective System (BREDS): Proposed Algorithm

Balanced Reputation Detective System (BREDS): Proposed Algorithm Balanced Reputation Detective System (BREDS): Proposed Algorithm Pallavi Bansal 1, Narender Singh 2 1,2Computer science &Engineering department, G.I.T.M, Bilaspur, Yamunanagar, Haryana ---------------------------------------------------------------------***---------------------------------------------------------------------

More information

Exploiting Local Popularity To Prune Routing Indices In Peer-to-Peer System

Exploiting Local Popularity To Prune Routing Indices In Peer-to-Peer System Exploiting Local Popularity To Prune Routing Indices In Peer-to-Peer System Stephane Bressan Department of Computer Science National University of Singapore steph@comp.nus.edu.sg Achmad Nizar Hidayanto

More information

Department of Computer Science Institute for System Architecture, Chair for Computer Networks. File Sharing

Department of Computer Science Institute for System Architecture, Chair for Computer Networks. File Sharing Department of Computer Science Institute for System Architecture, Chair for Computer Networks File Sharing What is file sharing? File sharing is the practice of making files available for other users to

More information

Characterizing Unstructured Overlay Topologies in Modern P2P File-Sharing Systems Technical Report CIS-TR University of Oregon

Characterizing Unstructured Overlay Topologies in Modern P2P File-Sharing Systems Technical Report CIS-TR University of Oregon Characterizing Unstructured Overlay Topologies in Modern P2P File-Sharing Systems Technical Report CIS-TR-25- University of Oregon June 3, 25 Daniel Stutzbach, Reza Rejaie University of Oregon {agthorr,reza}@cs.uoregon.edu

More information

Napster and Gnutella: a Comparison of two Popular Peer-to-Peer Protocols. Anthony J. Howe Supervisor: Dr. Mantis Cheng University of Victoria

Napster and Gnutella: a Comparison of two Popular Peer-to-Peer Protocols. Anthony J. Howe Supervisor: Dr. Mantis Cheng University of Victoria Napster and Gnutella: a Comparison of two Popular Peer-to-Peer Protocols Anthony J Howe Supervisor: Dr Mantis Cheng University of Victoria February 28, 2002 Abstract This article presents the reverse engineered

More information

Peer to peer networking: Main aspects and conclusions from the view of Internet service providers

Peer to peer networking: Main aspects and conclusions from the view of Internet service providers Peer to peer networking: Main aspects and conclusions from the view of Internet service providers Gerhard Haßlinger, Department of Computer Science, Darmstadt University of Technology, Germany Abstract:

More information

Efficient Content Location Using Interest-Based Locality in Peer-to-Peer Systems

Efficient Content Location Using Interest-Based Locality in Peer-to-Peer Systems Efficient Content Location Using Interest-Based Locality in Peer-to-Peer Systems Kunwadee Sripanidkulchai Bruce Maggs Hui Zhang Carnegie Mellon University, Pittsburgh, PA 15213 {kunwadee,bmm,hzhang}@cs.cmu.edu

More information

Super-Agent Based Reputation Management with a Practical Reward Mechanism in Decentralized Systems

Super-Agent Based Reputation Management with a Practical Reward Mechanism in Decentralized Systems Super-Agent Based Reputation Management with a Practical Reward Mechanism in Decentralized Systems Yao Wang, Jie Zhang, and Julita Vassileva Department of Computer Science, University of Saskatchewan,

More information

On Video Content Delivery in Wireless Environments

On Video Content Delivery in Wireless Environments , pp.81-85 http://dx.doi.org/10.14257/astl.2014.65.20 On Video Content Delivery in Wireless Environments Po-Jen Chuang and Hang-Li Chen Department of Electrical Engineering Tamkang University Tamsui, New

More information

PROPOSAL AND EVALUATION OF A COOPERATIVE MECHANISM FOR HYBRID P2P FILE-SHARING NETWORKS

PROPOSAL AND EVALUATION OF A COOPERATIVE MECHANISM FOR HYBRID P2P FILE-SHARING NETWORKS PROPOSAL AND EVALUATION OF A COOPERATIVE MECHANISM FOR HYBRID P2P FILE-SHARING NETWORKS Hongye Fu, Naoki Wakamiya, Masayuki Murata Graduate School of Information Science and Technology Osaka University

More information

HollyShare: Peer-to-Peer File Sharing Application

HollyShare: Peer-to-Peer File Sharing Application HollyShare: Peer-to-Peer File Sharing Application ICS 243A Class Project Songmei Han Bijit Hore Ilya Issenin Sean McCarthy Shannon Tauro HollyShare Project: Final Report 1 INTRODUCTION...3 SURVEY...5 Architecture

More information

Network Layer Performance in Peer-to-Peer File Sharing Systems

Network Layer Performance in Peer-to-Peer File Sharing Systems Network Layer Performance in Peer-to-Peer File Sharing Systems Raffaele Bolla, Roberto Bruschi, Michele Sciuto Dept of Communication, Computer and System Sciences, University of Genoa, Italy Email: raffaele.bolla,

More information

Adapting Distributed Hash Tables for Mobile Ad Hoc Networks

Adapting Distributed Hash Tables for Mobile Ad Hoc Networks University of Tübingen Chair for Computer Networks and Internet Adapting Distributed Hash Tables for Mobile Ad Hoc Networks Tobias Heer, Stefan Götz, Simon Rieche, Klaus Wehrle Protocol Engineering and

More information

A Utility Based Incentive Scheme for P2P File Sharing in Mobile Ad Hoc Networks

A Utility Based Incentive Scheme for P2P File Sharing in Mobile Ad Hoc Networks A Utility Based Incentive Scheme for P2P File Sharing in Mobile Ad Hoc Networks Afzal Mawji and Hossam Hassanein Telecommunications Research Lab, School of Computing Queen s University Kingston, Ontario,

More information

Exploiting Social Interactions in Mobile Systems

Exploiting Social Interactions in Mobile Systems Exploiting Social Interactions in Mobile Systems Andrew G. Miklas 1, Kiran K. Gollu 1,KelvinK.W.Chan 2,StefanSaroiu 1, Krishna P. Gummadi 3, and Eyal de Lara 1 1 University of Toronto 2 Google 3 MPI for

More information

FS2You: Peer-Assisted Semi-Persistent Online Storage at a Large Scale

FS2You: Peer-Assisted Semi-Persistent Online Storage at a Large Scale FS2You: Peer-Assisted Semi-Persistent Online Storage at a Large Scale Ye Sun, Fangming Liu, Bo Li, Baochun Li, Xinyan Zhang Hong Kong University of Science & Technology, University of Toronto, Roxbeam

More information

THE use of peer-to-peer (P2P) applications is growing dramatically,

THE use of peer-to-peer (P2P) applications is growing dramatically, IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 12, NO. 2, APRIL 2004 219 Analyzing Peer-To-Peer Traffic Across Large Networks Subhabrata Sen, Member, IEEE, and Jia Wang, Member, IEEE Abstract The use of peer-to-peer

More information

A Comparative Study of Tree-based and Mesh-based Overlay P2P Media Streaming

A Comparative Study of Tree-based and Mesh-based Overlay P2P Media Streaming A Comparative Study of Tree-based and Mesh-based Overlay P2P Media Streaming Chin Yong Goh 1,Hui Shyong Yeo 1, Hyotaek Lim 1 1 Dongseo University Busan, 617-716, South Korea cgnicky@gmail.com, hui_shyong@hotmail.com,

More information

Dynamic Search Algorithm used in Unstructured Peer-to-Peer Networks

Dynamic Search Algorithm used in Unstructured Peer-to-Peer Networks Dynamic Search Algorithm used in Unstructured Peer-to-Peer Networks B.SRIKANTH M.Tech Student QIS College of Engineering & Technology, Ongole, Prakasam Dist, AP, India. K.VENKATESWARA RAO Assoc. Prof in

More information

On the Impact of P2P File Sharing Traffic Restrictions on User Perceived Performance

On the Impact of P2P File Sharing Traffic Restrictions on User Perceived Performance On the Impact of P2P File Sharing Traffic Restrictions on User Perceived Performance Ricardo Lopes Pereira, Teresa Vazão Instituto Superior Técnico Av. Prof. Dr. Cavaco Silva, 2744-016 Porto Salvo, Portugal

More information

Peer-to-Peer Networks Organization and Introduction 1st Week

Peer-to-Peer Networks Organization and Introduction 1st Week Peer-to-Peer Networks Organization and Introduction 1st Week Department of Computer Science 1 Peer-to-Peer Networks Organization 2 2 Web & Dates Web page http://cone.informatik.uni-freiburg.de/lehre/vorlesung/

More information

New Structured P2P Network with Dynamic Load Balancing Scheme

New Structured P2P Network with Dynamic Load Balancing Scheme New Structured P2P Network with Dynamic Load Balancing Scheme Atushi TAKEDA, Takuma OIDE and Akiko TAKAHASHI Department of Information Science, Tohoku Gakuin University Department of Information Engineering,

More information

Denial of Service Resilience in Peer to Peer. D. Dumitriu, E. Knightly, A. Kuzmanovic, I. Stoica, W. Zwaenepoel Presented by: Ahmet Canik

Denial of Service Resilience in Peer to Peer. D. Dumitriu, E. Knightly, A. Kuzmanovic, I. Stoica, W. Zwaenepoel Presented by: Ahmet Canik Denial of Service Resilience in Peer to Peer File Sharing Systems D. Dumitriu, E. Knightly, A. Kuzmanovic, I. Stoica, W. Zwaenepoel Presented by: Ahmet Canik Outline 1. Background on P2P systems 2. File

More information

Transfer Rate Models for Gnutella Signaling Traffic

Transfer Rate Models for Gnutella Signaling Traffic Transfer Rate Models for Gnutella Signaling Traffic Dragos Ilie Dept. of Telecommunication Systems School of Engineering Blekinge Institute of Technology 371 79 Karlskrona, Sweden dragos.ilie@bth.se Adrian

More information

Measurement of edonkey Activity with Distributed Honeypots

Measurement of edonkey Activity with Distributed Honeypots Measurement of edonkey Activity with Distributed Honeypots Oussama Allali, Matthieu Latapy and Clémence Magnien LIP6 CNRS and Université Pierre et Marie Curie (UPMC Paris 6) 14, avenue du Président Kennedy

More information

Delft University of Technology Parallel and Distributed Systems Report Series. The Peer-to-Peer Trace Archive: Design and Comparative Trace Analysis

Delft University of Technology Parallel and Distributed Systems Report Series. The Peer-to-Peer Trace Archive: Design and Comparative Trace Analysis Delft University of Technology Parallel and Distributed Systems Report Series The Peer-to-Peer Trace Archive: Design and Comparative Trace Analysis Boxun Zhang, Alexandru Iosup, and Dick Epema {B.Zhang,A.Iosup,D.H.J.Epema}@tudelft.nl

More information

Kademlia: A Peer-to-peer Information System Based on the XOR Metric

Kademlia: A Peer-to-peer Information System Based on the XOR Metric Kademlia: A Peer-to-peer Information System Based on the XOR Metric Petar Maymounkov and David Mazières {petar,dm}@cs.nyu.edu http://kademlia.scs.cs.nyu.edu Abstract We describe a peer-to-peer system which

More information

A Comparison Study of Qos Using Different Routing Algorithms In Mobile Ad Hoc Networks

A Comparison Study of Qos Using Different Routing Algorithms In Mobile Ad Hoc Networks A Comparison Study of Qos Using Different Routing Algorithms In Mobile Ad Hoc Networks T.Chandrasekhar 1, J.S.Chakravarthi 2, K.Sravya 3 Professor, Dept. of Electronics and Communication Engg., GIET Engg.

More information

Search and Replication in Unstructured Peer-to-Peer Networks

Search and Replication in Unstructured Peer-to-Peer Networks Search and Replication in Unstructured Peer-to-Peer Networks Jacobs University 14-04-2008 Decentralized Unstructured P2P Decentralized Unstructured P2P Decentralized Unstructured P2P - no central server

More information

FS2You: Peer-Assisted Semi-Persistent Online Storage at a Large Scale

FS2You: Peer-Assisted Semi-Persistent Online Storage at a Large Scale FS2You: Peer-Assisted Semi-Persistent Online Storage at a Large Scale Ye Sun, Fangming Liu, Bo Li, Baochun Li, Xinyan Zhang Hong Kong University of Science & Technology, University of Toronto, Roxbeam

More information

Performance Evaluation of AODV, OLSR Routing Protocol in VOIP Over Ad Hoc

Performance Evaluation of AODV, OLSR Routing Protocol in VOIP Over Ad Hoc (International Journal of Computer Science & Management Studies) Vol. 17, Issue 01 Performance Evaluation of AODV, OLSR Routing Protocol in VOIP Over Ad Hoc Dr. Khalid Hamid Bilal Khartoum, Sudan dr.khalidbilal@hotmail.com

More information